[SPARK-3906][SQL] Adds multiple join support for SQLContext

Author: Cheng Lian <lian.cs.zju@gmail.com>

Closes #2767 from liancheng/multi-join and squashes the following commits:

9dc0d18 [Cheng Lian] Adds multiple join support for SQLContext

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/SqlParser.scala

@@ -181,9 +181,11 @@ class SqlParser extends AbstractSparkSQLParser {
     )
 
   protected lazy val joinedRelation: Parser[LogicalPlan] =
-    relationFactor ~ joinType.? ~ (JOIN ~> relationFactor) ~ joinConditions.? ^^ {
-      case r1 ~ jt ~ r2 ~ cond =>
-        Join(r1, r2, joinType = jt.getOrElse(Inner), cond)
+    relationFactor ~ rep1(joinType.? ~ (JOIN ~> relationFactor) ~ joinConditions.?) ^^ {
+      case r1 ~ joins =>
+        joins.foldLeft(r1) { case (lhs, jt ~ rhs ~ cond) =>
+          Join(lhs, rhs, joinType = jt.getOrElse(Inner), cond)
+        }
     }
 
   protected lazy val joinConditions: Parser[Expression] =

sql/core/src/test/scala/org/apache/spark/sql/SQLQuerySuite.scala

@@ -720,4 +720,15 @@ class SQLQuerySuite extends QueryTest with BeforeAndAfterAll {
     checkAggregation("SELECT key + 2, COUNT(*) FROM testData GROUP BY key + 1")
     checkAggregation("SELECT key + 1 + 1, COUNT(*) FROM testData GROUP BY key + 1", false)
   }
+
+  test("Multiple join") {
+    checkAnswer(
+      sql(
+        """SELECT a.key, b.key, c.key
+          |FROM testData a
+          |JOIN testData b ON a.key = b.key
+          |JOIN testData c ON a.key = c.key
+        """.stripMargin),
+      (1 to 100).map(i => Seq(i, i, i)))
+  }
 }